我正在为自己的兄弟情谊编写一个Django应用程序以对rushees进行投票,我正在尝试优化我的一个查询,该查询对票数进行计数,并打印出计数值以及来自其应用程序的信息。 Django调试工具栏告诉我我有很多重复查询
(为清晰起见,下面的代码已被截断和编辑)
models.py
votechoices = ((1, "Yes"),(2, "No"),(3, "Abstain"))
class Vote(models.Model):
brother = models.ForeignKey(Brother)
rushee = models.ForeignKey(Rushee)
choice = models.IntegerField(choices=votechoices, default=3)
class Rushee(models.Model):
first_name = models.CharField(max_length=40)
last_name = models.CharField(max_length=40)
#ETC, ETC
class Application(models.Model):
rushee = models.ForeignKey(Rushee)
address = models.CharField(max_length=200)
cellphone = models.CharField(max_length=30)
#ETC, ETC
views.py
def getvotecount(request):
# get all the applications ( we only vote on people who have an application)
applicationobjs = Application.objects.select_related('rushee').all()
# iterate through the applications and count the votes
for app in applicationobjs.iterator():
#>>>> This Query below is a seperate query everytime! So that means that If we have 200 rushees there are 200 queries!
counts = Vote.objects.filter(rushee=app.rushee).values('choice').annotate(count=Count('choice'))
votesfor = sum([x['count'] for x in counts if x['choice'] == 1])
votesagainst = sum([x['count'] for x in counts if x['choice'] == 2])
result = [app.rushee.first_name + ' ' + app.rushee.last_name,
app.address, app.cellphone,
str(votesfor), str(votesagainst),]
# Uninteresting stuff below that will group together and return the results
我正在尝试在(>>>>)标记的视图中优化查询,以使我可以返回每个rushee的票数,而无需每次都运行单独的查询!
附加信息:
sqlite后端,rushees的数量比应用程序的数量多,我们只对确实有应用程序的rushees进行投票
最佳答案
您可以使用conditional expressions在一个查询中完成所有操作:
from django.db.models import Case, IntegerField, Sum, When
rushees = Rushee.objects.annotate(
votes_for=Sum(
Case(
When(vote=1, then=1),
default=0,
output_field=IntegerField(),
)
),
votes_against=Sum(
Case(
When(vote=2, then=1),
default=0,
output_field=IntegerField(),
)
)
)
结果查询集中的
rushees
将具有votes_for
和votes_against
属性,并具有每个计数。假设没有针对没有申请的rushees
的投票,但如果有投票,您可以轻松地将其过滤掉。